Claude Fable 5 (xHigh) — benchmark results
Provider: Anthropic. Released 2026-06-09. Access: API.
Unified ELO 2074 ± 35, rank #6 of 1806 rated models, from 8 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| ARC-AGI-1 | 98.5 | Accuracy (%) | 99.7 |
| ARC-AGI-2 | 88.33 | Accuracy (%) | 97.7 |
| Epoch AI - ECI | 161.53 | ECI Score | 97.3 |
| Epoch AI - Cursorbench | 72 | Score | 96.7 |
| CursorBench 3.1 | 68.4 | Score (%) | 95.1 |
| LiveBench | 78.57 | LiveBench average (self-reported) | 92.9 |
| SimpleQA Verified | 68.3 | Accuracy (%) | 92.9 |
| FrontierCode | 46.3 | Main Score (self-reported) | 90 |
Interactive version: theaggregate.ai/model?slug=claude-fable-5-xhigh · How It Works · Data refreshed daily, snapshot 2026-08-07.